Liquid Metal-based Sticky Conductor for Wearable and Real-Time Electromyogram Monitoring with Machine Learning Classification

Abstract

Skin electronics face challenges related to the interface between rigid and soft materials, resulting in discomfort and signal inaccuracies. This study presents the development and characterization of a liquid metal-polydimethylsiloxane (LM-PDMS) sticky conductor designed for wearable electromyography (EMG) monitoring. The conductor leverages a composite of LM inks and PDMS, augmented with silver nanowires (AgNWs) and surface-modified with mercaptoundecanoic acid (MUD) to enhance conductivity. The mechanical properties of the PDMS matrix were optimized using Triton-X to achieve a flexible and adhesive configuration suitable for skin contact. Our LM-PDMS sticky conductor demonstrated excellent stretchability, could endure up to 300 % strain without damage, and maintained strong adherence to the skin without relative displacement. Biocompatibility tests confirmed high cell viability, making it suitable for long-term use. EMG signal analysis revealed reliable muscle activity detection, with advanced signal processing techniques effectively filtering noise and stabilizing the baseline. Furthermore, we employed machine learning algorithms to classify EMG signals, achieving high accuracy in distinguishing different muscle activities. This study showcases the potential of LM-PDMS sticky conductors for advanced wearable bioelectronics, offering promising applications in personalized healthcare and real-time muscle activity monitoring.